Embed Size (px)
Citation preview
Results you can rely on
Insights from Continuous Monitoring of LNAPL NSZD Rates
Fourth International Symposium on Bioremediation and
Sustainable Environmental Technologies
Miami, Florida
May 22-25, 2017
Keith Piontek, P.E. Tom Sale, PhDEric D. Emerson, EIT Kayvan Karimi Askarani
NSZD Rate Measurement Activities
• LNAPL Zone – Pipeline terminal in the Midwest
– Historic LNAPL accumulations > 1’ over a 10-acre area
– Weathered fuel mixture, dominated by gasoline and diesel
• Passive CO2 Flux Measurements– September 2012 through January 2015
– 7 multi-location events
• Biogenic Heat Monitoring– Phase 1 System became operational April 2014
– System modified and expanded in November 2016
– This presentation covers the 2.6 years before system modification
EPA Science Advisory BoardMay 2001
“MNA is a knowledge-based remedy…Instead of imposing active controls, as in engineered remediation, MNA involves understanding and documenting naturally occurring processes that reduce risks of exposure to acceptable levels.”
Hydrogeologic Setting
3
Fine-grained overbank deposits
Alluvial sands
Range of Typical Water Table Fluctuation
NSZD Conceptual Model
4Image Credit: Emily Stockwell/CSU
NSZD Rate Measurements –
Passive CO2 Flux Method
5
M-1
M-2
M-3
N-1
N-2
N-3
0
2,000
4,000
6,000
8,000
10,000
12,000
CO
2 F
lux
Rat
e (g
allo
ns
LNA
PL
per
acr
e-y
ear)
0
2,000
4,000
6,000
Running Average Event Average
Biogenic Heat Monitoring System
Observed Temperatures
• Elevated T (red v blue) reflects exothermic degradation of hydrocarbons
• T provides a signal to track extent of active NSZD and LNAPL
Credit: Emily Stockwell/CSU
NSZD Thermal Signature
BackgroundLNAPLCorrected TTT
LIF Data
368
Thermal Image Credit: Emily Stockwell/CSU
z
Z1
Z2
Figure 1. Energy balance volume consisting of the NAPL impacted area
x
y
Δz
Δx
Δy
𝐿𝑜𝑠𝑠𝑅𝑎𝑡𝑒 =−𝐸 𝑅𝑋𝑁∆𝐻𝑟
Estimating NSZD Rates
1 – 2 W/m2
.
Temperature Data Input
Energy Balance Model NSZD Rate Output
10
0
1
2
3
4
5
6
7
8
9
10
0
50
0
10
00
15
00
20
00
25
00
30
00
35
00
40
00
45
00
50
00
55
00
60
00
65
00
70
00
75
00
80
00
85
00
90
00
95
00
Nu
mb
er
of
Me
asu
rem
en
ts
Distribution of Rate Measurements (gallons per acre-year, 2 week average)
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
Apr-14 Oct-14 Apr-15 Oct-15 Apr-16 Oct-16
Monthly Loss Rate (gallons per acre per year)
0
2,000
4,000
6,000
8,000
10,000
Apr-14 Aug-14 Dec-14 Apr-15 Aug-15 Dec-15 Mar-16 Jul-16 Nov-16
Cumulative Loss (gallons per acre)
Method Comparison(All Measurements at 3 Locations Along a Transect)
11
Passive CO2 Flux Biogenic Heat
September 2012 through December 2014
April 2014 throughOctober 2016
Six events Continuous Monitoring
14C Correction for Background CO2
Cumulative Loss of 10,300 gallons per acre
Average: 4,900 gallonsper acre per year
Average: 4,000 gallonsper acre per year
What is the Cause of Temporal Variability?
Seasonal Pattern Potential Cause
• Temperature – Proven relationship between
temperature and rate
• Water Table Fluctuations– Hypothesis of faster rate in
unsaturated portion of smear zone
• Soil Moisture– Both gas diffusion and
advection a function of air-filled porosity
12
0
1000
2000
3000
4000
5000
6000
7000
Jan
(2
)
Feb
(2
)
Mar
(2
)
Ap
r (3
)
May
(3
)
Jun
(3
)
Jul (
3)
Au
g (3
)
Sep
t (3
)
Oct
(3
)
No
v (2
)
Dec
(2
)
Ave
rage
Rat
e (g
allo
ns
per
acr
e p
er y
ear)
Month
Average Rate by Month
Background Temperature Over Time at Various Depths
13
0
5
10
15
20
25
30
Mar-14 Jun-14 Sep-14 Dec-14 Apr-15 Jul-15 Oct-15 Jan-16 May-16 Aug-16 Nov-16
Tem
per
atu
re (
Cel
ciu
s)
1' bgs
10' bgs
19' bgs
August (1’)
October (10’)
December (19’)
NSZD Rate vs
Background Soil Temperature (19’ bgs)
14
Is the assumption of a 5 month lag reasonable?• Bemidji Site (Sihota, et. al., 2016): Yes• Site-specific assessment: No
Water Table Fluctuation
15
716
718
720
722
724
726
728
730
732
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
Date
Gro
un
dw
ater
Ele
vati
on
(ft
. AM
SL)
Loss
Rat
e
Avg Loss Rate_Sto GW-14 GW Elevation
Higher rate with more of the LNAPL zone submerged
Rate vs Annual Soil Moisture Cycle
16
Derivation of monthly soil dryness factor based on long-term averages of:• Pan evaporation• Precipitation
Findings for Terminal Site
• Engineering: the biogenic heat monitoring
system worked, and its still going
• NSZD Rate (northern transect):
– CO2 flux method: 4,900 gallons per acre per year
– Thermal method: 4,000 gallons per acre per year
– Validation of the CSU energy balance model for
rate derivation
• Regulatory:
– NSZD is the approved remedy for LNAPL zone and
aqueous phase plume
– Continued tracking of LNAPL mass loss and plume
status required17
18
“Limitation” of the Biogenic Heat Monitoring Method
TRC’s Experience with the CSU System
Background correction needed. Yes…but not that hard.
Top and bottom of oxidation zone must be known.
Not with any precision.
Downward heat flux an uncertainty Easily accounted for.
Must be careful of thermal anomalies. Yes!
Unless the site is in an equatorial zone, need monitoring throughout the year.
A system for continuous monitoring exists.
Its complicated - have to account for the heat generated by all biological reactions, variable heat production.
Its not that complicated to derive a rate that has less uncertainty than other methods.
Findings for Broader Consideration
19
• Rate measurements based on
biogenic gas efflux at the
ground surface
– Significant temporal variability
– Cause of variability?
• Biogenic heat monitoring
approach
– Continuous monitoring
accounts for temporary
variability
– The hardware and energy
balance model has been
validated
– More cost-effective than
surface efflux methods for:
• “True annual average”
LNAPL mass loss
• Long-term monitoring
Questions?
Kayvan Karimi AskaraniP: 970.213.7419E: [email protected]
Tom Sale, PhDP: 970.491.8413E: [email protected]
Eric D. EmersonP: 970.484.3262 ext. 15968E: [email protected]
Keith Piontek, PEP: 314.241.2694 x 11 M: 314.882.6531E: [email protected]
